Manufacturing

Highlights The global fight against the Delta variant of COVID-19 continued to show progress in the month of September, but not without cost. Growth in services activity slowed meaningfully, which has likely delayed the return to potential output in the US until March of next year (at the earliest). However, even with this revised timeline, maximum employment remains a very possible outcome by next summer, barring a further extension of the pandemic in advanced economies. In this regard, the Fed’s likely decision at its next meeting to taper the rate of its asset purchases makes sense and is consistent with a first rate hike in the second half of 2022. The rise in long-maturity bond yields following this month’s Fed meeting is consistent with the view that 10-year Treasurys are overvalued and that yields will trend higher over the coming year. Fixed-income investors should stay short duration. The degree to which global shipping costs are being driven by the forces of supply versus demand will affect the Fed's criteria for liftoff next year, via changes in goods prices as well as consumer expectations for inflation. In our view, a detailed examination of shipping prices over the past 18 months points to a future pace of inflation that is not dangerously above-target, but does meet the Fed’s liftoff criteria. A mix-shift in consumer spending, away from goods and toward services, is not a threat to economic activity or S&P 500 earnings – so long as the decline in the former is not outsized relative to the rise in the latter. It will, however, disproportionately impact China, and could be the trigger for meaningful further easing by Chinese policymakers. In the interim, a catalyst for EM stocks may remain elusive. We continue to recommend an overweight stance toward value versus growth stocks and global ex-US versus the US, particularly in favor of developed markets ex-US. Investors should remain cyclically overweight stocks versus bonds, although it is possible that both assets will post negative returns for a short period at some point over the coming 12 months in response to higher long-maturity bond yields. Still, we expect both stock prices and the stock-to-bond ratio to be higher a year from today. Feature The global fight against the Delta variant of COVID-19 continued to show progress in the month of September. Chart I-1 highlights that an estimate of the reproduction rate of the disease in developed economies has fallen below one, and the weekly change in hospitalizations in both the US and UK – the two countries at the epicenter of the Delta wave that have not reintroduced widespread COVID-19 control measures – have fallen back into negative territory. In addition, we estimate that approximately 6% of the world’s population received vaccines against COVID-19 in September, with now 45% of the globe having received a first dose and 33% now fully vaccinated. Pfizer’s announcement last week that it has found a “favorable safety profile and robust neutralizing antibody responses” from its vaccine trial in children five to eleven years of age suggests that the FDA may grant emergency use authorization within weeks, which would likely raise the vaccination rate in the US (and ultimately other advanced economies) by at least 5 percentage points in fairly short order. This would also further reduce the impact of school/classroom closures on the labor market, via both an increased participation rate and increased hiring in the education sector. This fight, however, has not been without cost. US jobs growth slowed significantly in August, manufacturing and services PMIs continued to slow in September, and, as Chart I-2 highlights, the normalization in transportation use that was well underway in the first half of the year has clearly inflected in both the US and UK in response to the spread of Delta. Consensus market expectations for Q3 growth have been cut in the US, and to a lesser extent in the euro area, and the Fed reduced its forecast for 2021 real GDP growth from 7% to 5.9% following the September FOMC meeting. Chart I-1The Delta Wave Continues To Abate... Chart I-2...But At A Cost To Economic Activity   The Path Toward Eventually Tighter Monetary Policy It has been surprising to some investors that the Fed has moved forward with their plans to taper the rate of its asset purchases against this backdrop of slowing near-term growth – an event that now seems likely to occur at its next meeting barring a disastrous September payroll report. In our view, this is not especially surprising, given that the Fed has expressed a desire for net purchases to reach zero before they raise interest rates for the first time. Chair Powell noted during last week’s press conference that FOMC participants felt a “gradual tapering process that concludes around the middle of next year is likely to be appropriate”, underscoring that the Fed wants the flexibility to raise interest rates in the second half of next year. The timing of the first Fed rate hike is entirely subject to the evolution of the economic data over the next year, and is not, in any way, calendar-based. But we presented in last month's Special Report why the Fed’s maximum employment criteria may be met as early as next summer,1 and the Fed’s projections for the pace of tapering are consistent with our analysis. Chart I-3Maximum Employment Remains A Very Possible Outcome By Next Summer The Fed’s most recent Summary of Economic Projections (“SEP”) also seemingly confirmed Fed Vice Chair Richard Clarida’s view that a 3.8% unemployment rate is consistent with maximum employment, barring any issues with the “breadth and inclusivity” of the labor market recovery. We noted in last month’s report that these issues are unlikely in a scenario where jobs growth is sufficiently high to bring down the unemployment rate below 4%. Chart I-3 highlights that both the Fed’s forecast and Bloomberg consensus expectations imply a closed output gap by March, even after factoring in the near-term impact of the Delta variant. Consequently, maximum employment remains a very possible outcome by next summer, barring a further extension of the pandemic in advanced economies. Long-maturity bond yields rose following the Fed meeting, which is also not especially surprising given how low yields have fallen relative to the fair value implied by the Fed’s SEP forecasts even assuming a December 2022 initial rate hike. Chart I-4 highlights that the fair value of the 10-year Treasury yield today is roughly 2% using this approach, rising to 2.15% by next summer. Ironically, the September SEP update modestly lowered the fair value shown in Chart I-4 relative to what would otherwise have been the case, as it implied that the Fed is expecting to raise interest rates at a pace of approximately three hikes per year – rather than the four that prevailed prior to the pandemic. Investors should also note that the fair value for the 10-year yield is nontrivially lower based on market participant and primary dealer estimates of the terminal Fed funds rate (also shown in Chart I-4), although they still imply that long-maturity yields should trend higher over the coming year. Global Trade, Inflation, And The Fed A return to maximum employment will likely signal the onset of monetary policy tightening, as long as the Fed's inflation criteria for liftoff have been met. For now, inflation is signaling a green light for hikes next year, even after excluding the prices of COVID-impacted services and cars (Chart I-5). In fact, more recently, CPI ex-direct COVID effects has been pointing in the “non-transitory” direction, which continues to prompt questions from investors about whether the Fed will be forced to hike earlier than it currently expects for reasons other than a return to maximum employment. Chart I-4US Long-Maturity Bond Yields Are Set To Move Higher Over The Coming Year Chart I-5For Now, Inflation Is Signaling A Green Light For Hikes Next Year   At least some portion of the current pace of increase in consumer goods prices is tied to surging import costs, which have run well in-excess of what would be predicted by the relationship with the US dollar (Chart I-6). This, in turn, is being driven by an explosion in shipping costs that has occurred since the onset of the pandemic, which is being driven both by demand and supply-side factors (Chart I-7). Chart I-6US CPI Is Being Affected By Surging Import Prices... Chart I-7...Which Are Being Driven By An Explosion In Shipping Costs   The degree to which global shipping costs are being driven by the forces of supply versus demand will affect the Fed's criteria for liftoff next year, via changes in goods prices as well as consumer expectations for inflation. To the extent that demand side factors are mostly responsible, investors should have higher confidence that the recent surge in consumer prices is transitory, because a shift away from above-trend goods spending and toward below-trend services spending is likely over the coming year. If supply-side factors are mostly responsible, then it is conceivable that the global supply chain impact on consumer goods prices will persist for longer than would otherwise be the case, potentially raising the odds of a larger or more sustained rise in inflation expectations. In our view, a detailed examination of shipping prices over the past 18 months points to a mix of both demand and supply effects, even since the beginning of 2021. However, as we highlight below, several facts point toward the view that supply-side factors will be the dominant driver over the coming year, and that they are more likely to exert a disinflationary/deflationary rather than inflationary effect: Chart I-8 breaks down the cumulative change in the overall Freightos Baltic Index by route since December 2019. The chart makes it clear that shipping costs from China/East Asia to the West Coast of the US have risen far more than any other route, underscoring that US demand for goods has been an important part of the rise in shipping costs. Chart I-8US Demand For Goods Is An Important Part Of The Shipping Cost Story Chart I-9US Goods Spending Has Clearly Been Boosted By US Fiscal Policy Chart I-9 shows the level of real US personal consumption expenditures on goods relative to its pre-pandemic trendline, underscoring both that goods spending is currently well-above trend, and that there have been two distinct phases of rising goods spending: from May to October 2020 following the passage of the CARES act, and from January to March 2021 following the December 2020 extension of UI benefits and in anticipation of the passage of the American Rescue Plan. Since March, US real goods spending has trended lower, a pattern that we expect will continue over the coming year. Chart I-10 highlights that while the global supply chain struggled heavily last year in response to surging demand and the lagging effects of labor shortages and factory shutdowns during the earliest phase of the pandemic, there were some signs of supply-side normalization in the first half of 2021. The chart highlights that the number of ships at anchor at the Los Angeles and Long Beach ports declined meaningfully from February to June, and global shipping schedule reliability tentatively improved in March. The chart also shows that shipping costs from China/East Asia to the West Coast of the US continued to rise in Q2 seemingly as a lagged response to the Jan-Mar rise in goods spending, but they were still low at the end of June compared to today’s levels. Chart I-10Supply-Side Factors Seem To Have Driven A Majority Of This Year's Increase In Shipping Costs In Q3, circumstances drastically changed. Shipping costs between China/East Asia to the West Coast of the US rapidly doubled, and the number of ships at anchor at the LA/LB ports exploded well past its peak in early February. This rise in China/US shipping costs since late-June has accounted for nearly 60% of the cumulative rise since the pandemic began, and cannot be attributed to increased demand. Instead, the increase in prices and the surge in port congestion in Q3 appears to have been caused by the one-month closure of the Port of Yantian that began in late-May, in response to an outbreak of COVID-19 in Guangdong province. Yantian is the fourth largest port in the world and exports a sizeable majority of global electronics given its close proximity to Shenzhen, underscoring the impact that its closure likely had on an already bottlenecked logistical system. There are two key points emanating from our analysis of global shipping costs. First, demand has been an important effect driving costs higher, but it does not appear to have driven most of the increase in shipping costs this year. Still, over the coming year, goods demand in advanced economies is likely to wane as consumer spending shifts from goods to services spending, which will help ease clogged global trade channels and lower shipping costs. Second, the (brief) evidence of supply-side normalization in the first half of 2021, when consumer demand was actually strengthening, suggests that the supply-side of the global trade system will turn disinflationary over the coming year if further COVID-related labor market shocks can be avoided. What does this mean for the Fed and the prospect of monetary policy tightening next year? In our view, the combination of a positive output gap, stable but normalized inflation expectations, and disinflation (or outright deflation) in COVID-related goods and services (including import prices) is likely to lead to a pace of inflation that meets the Fed’s liftoff criteria. Chart I-11 highlights that important longer-term inflation expectations measures have recently been well-behaved, despite a surge in actual inflation and shorter-term expectations for inflation. Aided by disinflation/deflation in certain high-profile COVID-related goods and services prices, this argues against meaningful upside risks to inflation. However, the current level of long-term expectations and the fact that the output gap is set to turn positive in the first half of next year argues against the notion that inflation will fall below target outside of COVID-related effects. As such, we continue to expect that the Fed will raise interest rates next year, potentially as early as next summer, driven by the progress towards maximum employment. Spending Shifts And The Equity Market We noted above, and in previous reports, that consumer spending in advanced economies is likely to continue to shift away from goods and toward services over the coming year. This raises the question of whether a contraction in goods spending will weigh disproportionately on the economy and equity earnings, given the close historical correlation between manufacturing activity and the business cycle. Chart I-12 illustrates this risk: in a hypothetical scenario in which real goods spending were to return to the trendline shown in Chart I-9 by March of next year, it would contract on the order of 10% on a year-over-year basis, on par with what occurred last year and vastly in excess of what even normally occurs during a recession. Chart I-11Longer-Term Inflation Expectations Remain Well-Behaved Chart I-12A Contraction In Goods Spending Is Likely Over The Coming Year   Chart I-12 is a hypothetical scenario and not a forecast, as there is some evidence that consumers are currently deferring durable goods purchases on the expectation that prices will become more favorable. In addition, a positive output gap next year implies that goods spending may settle above its pre-pandemic trendline. Nevertheless, the prospect of a potentially significant slowdown in goods spending has unnerved some investors, even given the prospect of improved services spending. Chart I-13highlights that this fear is understandable given how the US economy normally behaves. The top panel of the chart shows the year-over-year contribution to real GDP growth from real goods and services spending, and the bottom panel shows these contributions in absolute terms to better illustrate their relative magnitudes. The chart makes it clear that goods spending is normally a more forceful driver of economic activity than is the case for services spending, which ostensibly supports concerns that a significant slowdown in the former may be destabilizing for overall activity. Chart I-13Normally, Goods Spending Predominantly Drives Activity. Not This Cycle. However, Chart I-13 also highlights that the magnitude of the recent contribution to growth from services spending has been absolutely unprecedented in the post-WWII economic environment. This is not surprising given the nature of the COVID-19 pandemic, but it is important because it underscores that investors should not rely excessively on typical rules of thumb about how modern economies tend to function over the course of the business cycle. In terms of the impact on overall economic activity, investors should focus on the net impact of goods plus services spending. It is certainly possible that the former will slow at a pace that is not fully compensated by the latter, but our sense is that this is not likely to occur barring a further extension of the pandemic in advanced economies. Chart I-14Over The Past 5 Years, S&P 500 Sales Have Been More Correlated With Services Than Goods Spending Chart I-14 presents a similar conclusion for the US equity market. The chart highlights the historical five-year correlation between the quarterly growth of nominal spending and S&P 500 sales per share. The chart shows that S&P 500 revenue was more sensitive to goods versus services spending prior to the 1990s, when the US was more manufacturing-oriented and goods were more likely to be produced domestically than is the case today. Another gap in the correlation emerged following the global financial crisis when the US household sector underwent several years of deleveraging. But over the past five years, Chart I-14 highlights that S&P 500 revenue growth has actually been more strongly correlated with US services spending than goods spending. Some of this increased correlation might reflect technology-related services spending which could suffer in a post-pandemic environment, but the bottom line from Chart I-14 is that there is not much empirical support for the view that US equity fundamentals will be disproportionately impacted by a slowdown in goods spending, so long as services spending rises in lockstep. China: Exacerbating An Underlying Trend Chart I-15China Will Be Disproportionately Affected By Slowing DM Goods Spending China, on the other hand, will be disproportionately affected by slower goods spending in advanced economies, because its exports have disproportionately benefited from the surge in spending on goods over the past year. Chart I-15 highlights that Chinese export volume growth has exploded this year, and that current export growth is running at a pace of 10% in volume terms – significantly higher than has been the case on average over the past decade. Several problems in China have been in the headlines over the past few months: a regulatory crackdown by Chinese authorities on new economy companies, the situation with Evergrande and, more recently, power shortages that have forced factories in several key manufacturing hubs to curtail production as a result of China’s ban on coal imports from Australia (Chart I-16). However, the key point for investors is that these are not truly new risks to China’s growth outlook; rather, they are developments that have the potential to magnify the impact of an already established trend: the ongoing slowdown in China’s economy that has clearly been caused by a decline in its credit impulse (Chart I-17). In turn, China’s decelerating credit impulse has been caused by tighter regulatory and monetary policy. Chart I-16Power Outages: The Latest Negative Headline From China Chart I-17China Is Slowing Because Policymakers Have Tightened   BCA’s China Investment Strategy service has provided a detailed analysis of the ongoing Evergrande saga.2 In short, our view is that the government will likely restructure Evergrande’s debt to prevent the company’s crisis from evolving into a systemic financial risk. As such, Beijing may rescue the stakeholders of Evergrande, but likely not its shareholders. However, in terms of stimulating the broader economy, it is still not clear that Chinese policymakers are willing to engage in more than gradual or piecemeal stimulus, given a higher pain threshold for a slower economy and a lower appetite for leverage. This may change once Chinese export growth slows in response to a shift in DM spending from goods to services, as policymakers will no longer be able to rely on the external sector for support. This potentially offsetting nature of eventual Chinese stimulus and global goods spending underscores both the importance of a normalization in DM services spending as an impulse for global growth, as well as the fact that a catalyst for EM stocks may remain elusive over the tactical horizon. Investment Conclusions In Section 2 of this month’s report, we explain why the performance of US stocks may be flat versus their global peers over a structural time horizon. We also highlighted that US stocks are likely to earn low annualized total returns over the coming 10 years (between 1.8 - 4.7%), which would fall well short of the absolute return goals of many investors. Chart I-18Losses From Both Stocks And Bonds Are Rare, But Are Linked To Higher Rates Over the coming 6-12 month time horizon, we continue to recommend an overweight stance towards value vs. growth stocks and global ex-US vs. US, particularly in favor of developed markets ex-US. The relative performance of value vs. growth stocks is likely to benefit from the transition to a post-pandemic state and a rise in long-maturity bond yields, as monetary policy shifts towards the point of tightening. Regional equity trends have been closely correlated with style over the past two years, and the underperformance of growth strongly implies US equity underperformance. From an asset allocation perspective, investors should remain overweight stocks versus bonds over the coming year, although it is possible that both assets will post negative returns for a short period at some point over the coming 12 months. Chart I-18 highlights that outside of the context of recessions, months with negative returns from both stocks and long-maturity bonds are quite rare, but they tend to be associated with periods of monetary policy tightening (or in anticipation of such periods). Fundamentally, we do not see a rise in bond yields to any of the levels shown in Chart I-4 as being threatening to economic growth or necessarily implying lower equity market multiples. But the speed of adjustment in bond yields could unnerve equity investors, and there are open questions as to how far the equity risk premium can fall before T.I.N.A. – “There Is No Alternative” – becomes a less persuasive argument. As such, we would not rule out a brief correction in stocks at some point over the coming several months, but we expect both stock prices and the stock-to-bond ratio to be higher a year from today. Jonathan LaBerge, CFA Vice President The Bank Credit Analyst September 30, 2021 Next Report: October 28, 2021 II. The “Invincible” US Equity Market: The Longer-Term Outlook For US Stocks In Relative And Absolute Terms Since 2008, US equity outperformance versus global ex-US stocks has not been driven by stronger top-line growth. Instead, it has been caused by a narrowly-based increase in profit margins, the accretive impact of share buybacks on the EPS of US growth stocks, and an outsized expansion in equity multiples. To a lesser extent, the dollar has also boosted common currency relative performance. There are significant secular risks to these sources of US equity outperformance over the past 14 years. Elevated tech sector profit margins are likely to lead to increased competition and higher odds of regulatory action, leveraging has reduced the ability of US companies to continue to accrete EPS through changes to capital structure, relative multiples are not justified by relative ROE, and the US dollar is expensive and is likely to fall over a multi-year horizon. In absolute terms, we forecast that US stocks will earn annualized nominal total returns of between 1.8 - 4.7% over the coming decade, assuming 4-5% annual revenue growth, flat profit margins, a constant 2% dividend yield, and a constant equity risk premium. Long-maturity bond yields are below their equilibrium levels and are likely to rise in real terms over time, which will weigh on elevated equity multiples. Over the coming 6-12 months, our view that US 10-Year Treasury yields are likely to rise argues for an underweight stance toward growth versus value stocks. In turn, this implies that US stocks will underperform global stocks, especially versus developed markets ex-US. The risks that we have highlighted to the sources of US outperformance suggest that US stocks may be flat versus their global peers over the long-term, arguing for a neutral strategic allocation. It also suggests that investors should be prepared to accept more volatility in order to reduce the gap between expected and desired returns, and should look towards riskier investments and asset classes (such as real estate and alternative investments) as potential portfolio return enhancements. Chart II-1The US Has Massively Outperformed Other Equity Markets Since The Global Financial Crisis The US equity market has vastly outperformed its peers since the 2008/2009 global financial crisis. Chart II-1 highlights that an investment in US stocks at the end of 2007 is now worth over 4 times the invested amount, versus approximately 1.6 times for global ex-US stocks (when measured in US dollar terms). The chart also shows that USD-denominated total returns have been roughly the same for developed markets ex-US as they have been for emerging markets, highlighting the exceptional nature of US equities. In this report we provide a deep examination of the sources of US equity performance, their likely sustainability, and what this implies for long-term investor return expectations. US stocks have not outperformed because of stronger top-line (i.e. revenue) growth, and instead have benefitted from a narrowly-based increase in profit margins, active changes to capital structure that have benefitted stockholders, an outsized expansion in equity multiples relative to global stocks, and a structural appreciation in the US dollar. We conclude that there are significant risks to all of these sources of outperformance, and that a neutral strategic allocation to US equities is now likely warranted. We also highlight that, while a strategic overweight stance is still warranted toward stocks versus bonds, investors should no longer count on US stocks to deliver returns that are in line with or above commonly-cited absolute return expectations. This argues for a greater tolerance of volatility, and the pursuit of riskier investments and asset classes (such as real estate and alternative investments) as potential portfolio return enhancements. A Deep Examination Of US Outperformance Since 2008 Breaking down historical total return performance is the first step in judging whether US equities are likely to outperform their global ex-US peers on a structural basis. Below we deconstruct US and global total return performance over the past 14 years into six different components, and analyze the impact of some of these components on a sector-by-sector basis. The six components presented are: Total revenue growth for each equity market, in local currency terms The change in profit margins The impact of changes in capital structure and index composition The change in the trailing P/E ratio The income return from dividends The impact of changes in foreign exchange The sum of the first three factors explains the total growth in earnings per share over the period, and the addition of the fourth factor explains each market’s local currency price return. Income returns are added to explain total return over the period, with the sixth factor then explaining common currency total return performance. The FX effect for US stocks is zero by construction, given that we measure common currency performance in US$ terms. Chart II-2Strong US Returns Have Not Been Due To Strong Top Line Growth Chart II-2 presents the annualized absolute impact of these factors for the MSCI US index since 2008. The chart highlights that U.S. stock prices have earned roughly 11% per year in total return terms over the past 14 years, with significant contributions from revenue growth, multiple expansion, margins, and the return from dividends. Interestingly, however, Chart II-3 highlights that US equities have not significantly outperformed on the basis of the first factor, total local currency revenue growth, at least relative to overall global ex-US stocks (see Box II-1 for more details). DM ex-US stocks have experienced very weak revenue growth since 2008, but this has been compensated for by outsized EM revenue growth. It is also notable that US revenue growth has actually underperformed US GDP growth over the period, dispelling the notion that US equity outperformance has been due to strong top-line effects. Chart II-3The US Has Outperformed Due To Margins, Capital Structure, Multiples, And The Dollar Box II-1 Proxying The Impact Of Changes In Shares Outstanding We proxy the impact of changes in shares outstanding (and thus the impact of equity dilution / accretion) by dividing each index’s market capitalization by its stock price. This measure is not a perfect proxy, as changes in index composition (such as the addition/deletion of index constituents) will change the index’s market capitalization but not its stock price. We also calculate total revenue for each market by multiplying local currency sales per share by the market cap / stock price ratio, meaning that the total revenue growth figures shown in Chart II-3 should best be viewed as estimates that in some cases reflect index composition effects. However, Chart II-B1 highlights that adjusting the market cap / stock price ratio for the number of firms in the index does not meaningfully change our overall conclusions. This approach would imply a larger dilution effect for DM ex-US than suggested in Chart II-3, and a smaller effect for emerging markets (due to a significant rise in the number of EM index constituents since 2008). In addition, global ex-US revenue growth is modestly lower than US revenue growth when using this approach. But this gap would account for a fraction of US equity outperformance over the period, underscoring that the US has massively outperformed global ex-US stocks due to margin, capital structure, and multiple expansion effects. Chart II-B1The US Has Not Meaningfully Outperformed Due To Revenue Growth, No Matter How You Slice It Chart II-3 also highlights that global ex-US stocks have modestly outperformed the US in terms of the fifth factor, the income return from dividends. This has almost offset the negative FX return (the sixth factor) from a net rise in the US dollar over the period. What is clear from the chart is that the second, third, and fourth factors explain almost all of the difference in total return between US and global ex-US stocks since 2008. The US experienced a significant increase in profit margins versus a modest contraction for global ex-US, a modest fillip from changes in capital structure and index composition versus a substantial drag for ex-US stocks, and a sizable rise in equity multiples that has outpaced what has occurred around the globe in response to structurally lower interest rates. Chart II-4US Margin Outperformance Has Been Narrowly-Based The significant rise in aggregate US profit margins over the past 14 years has often been attributed to the strong competitiveness of US companies, but Chart II-4 highlights that the aggregate change mostly reflects a narrow sector composition effect. The chart shows the change in US and global ex-US profit margins by level 1 GICS sector since 2008, and underscores that overall profit margins outside of the US have fallen mostly due to lower oil prices. Conversely, in the US, profit margins have substantially risen in only three out of ten sectors: health care, information technology, and communication services. Chart II-5 highlights that global ex-US equity multiples have risen in a majority of sectors since 2008, but not by the same magnitude as what has occurred in the US. De-rating in the resource sector partially explains the gap, but stronger US multiple expansion in the heavily-weighted consumer discretionary, information technology, and communication services sectors appears to explain most of the gap in multiple expansion. Chart II-5Multiples Have Risen Globally, But More So For Broadly-Defined US Tech Stocks Finally, Charts II-6 & II-7 highlights that there has been a strong growth versus value dimension to the impact of changes in capital structure and index composition on regional equity performance. The charts show that equity dilution and other changes to index composition have caused a similar drag on the returns from value stocks in the US and outside the US. However, the charts also highlight that the more important effect has been the accretive impact of share buybacks on the EPS of US growth stocks, which has not been matched by growth stocks outside of the US. As noted in Box II-1, part of this gap may be explained by an increase in the number of companies included in the MSCI Emerging Markets index, but Chart II-8 highlights that the global ex-US ratio of market capitalization to stock price has still risen significantly over the past 14 years, in contrast to that of the US even after controlling for the number of index components. Chart II-6There Has Been A Strong Style Dimension… Chart II-7…To The Impact Of Changes In Capital Structure And Index Composition Chart II-8The Accretive Impact Of US Growth Stock Buybacks Has Not Been Matched Globally The bottom line for investors is that there have been multiple factors contributing to US equity outperformance since 2008, but aggregate top-line growth has not been one of them. Broadly-defined technology companies (including media & entertainment and internet retail firms) have been responsible for nearly all of the relative rise in profit margins and most of the relative expansion in multiples over the past 14 years, and US growth stocks have benefitted from the accretive impact of share buybacks to a larger degree than what has occurred globally. The Relative Secular Return Outlook For US Stocks We present below several structural risks to the continued outperformance of US equities for the factors that have been most responsible for this performance over the past 14 years. In some cases, these risks speak to the potential for US outperformance to end, not necessarily that the US will underperform. But even the cessation of US outperformance along one or more of these factors would be significant, as it would imply a potential inflection point in the most consequential trend in regional equity performance since the 2008/2009 global financial crisis. Profit Margins Chart II-9 presents the 12-month trailing combined profit margin for the US consumer discretionary, information technology, and communication services sector versus that of the remaining sectors. The chart underscores the points made by Chart II-4 in time series form, namely that the net increase in overall US profit margins since 2008 has been narrowly based. Chart II-9The US Profit Margin Expansion Has Been Driven By Broadly-Defined Tech Stocks Over a 6-12 month time horizon, the clear risk to US profit margins is an end to the COVID-19 pandemic. The profitability of broadly-defined tech stocks has surged during the pandemic, in response to a significant shift toward online goods purchases and elevated spending on tech equipment. A durable end to the pandemic is likely to reverse some of these spending patterns, which will likely weigh on margins for broadly-defined tech stocks. Chart II-10The Regulatory Risks Facing Big Tech Are Real Over the longer term, the risk is that extremely elevated profit margins are likely to increase the odds of regulatory action from Washington and invite competition. On the former point, our US Political Strategy service has highlighted that a bipartisan consensus in public opinion holds that Big Tech needs tougher regulation (Chart II-10), and this consensus grew substantially over the controversial 2020 political cycle.3 This regulatory pressure is currently best described as a “slow boil,” as not all surveys show strong majorities in favor of regulation, and Republicans and Democrats disagree on the aims of regulation. But the bottom line is that Big Tech is likely to remain in the hot seat after the various controversies of the pandemic and 2016-2020 elections, just as big banks faced tougher regulation in the wake of the subprime mortgage crisis. This underscores that a “slow boil” may turn into a faster one at some point over the secular horizon, which would very likely weigh on profit margins. Elevated tech sector profit margins makes regulatory action more likely because policymakers will perceive a stronger ability for these firms to weather a “regulatory shock.” On the latter point about competition, it is true that broadly-defined tech stocks follow a “platform” business model that will be difficult to supplant. These companies benefit from powerful network effects that have taken years to accrue, suggesting that they will not be rapidly replaced by competitors. Still, the experience of Microsoft in the years following its meteoric rise in the second half of the 1990s provides a cautionary tale for broadly-defined tech stocks today. In the late-1990s, it was difficult for investors to envision how Microsoft’s near-total product dominance of the PC ecosystem could ever be displaced, but it eventually lost market share due to the rise of mobile devices and their competing operating systems. In addition, Microsoft’s fundamental performance suffered even before the rise of the modern-day smartphone & mobile device market. Chart II-11 highlights the annualized components of Microsoft’s price return from 1999-2007 versus the late-1990s period, which underscores that changes in margins, changes in multiples, and stock price returns may be persistently negative in a scenario in which revenue growth slows (even if revenue growth itself remains positive). Chart II-11Microsoft Offers A Cautionary Tale For Dominant Business Models Some of the reversal of Microsoft’s fortunes during this period were self-inflicted, and the firm also suffered from an economy-wide slowdown in tech equipment spending as a result of the 2001 recession that persisted into the early years of the subsequent recovery. But the key point for investors is that company and sector dominance may wane, and the fact that broadly-defined tech sector profit margins are extremely elevated raises the risk that further increases may not materialize. Capital Structure And Index Composition As noted above, the beneficial impact from changes in capital structure and index composition for US equities has occurred due to the accretive impact of share buybacks on the EPS of US growth stocks, which has not been matched by growth stocks outside of the US. In our view, this accretive impact has occurred for two reasons. First, US growth stocks have taken advantage of historically low interest rates and leverage to shift their capital structure to be more debt-focused over the past 14 years. Second, this shift has been aided by the fact that US growth stocks have experienced stronger cash flows than their global peers, which have been used to service higher debt payments. However, Charts II-12 and II-13 suggest that this process may be in its late innings. Chart II-12 highlights that the US nonfinancial corporate sector debt service ratio (DSR) did indeed fall below that of the euro area following the global financial crisis, but that this reversed in 2016. At the onset of the pandemic, the US nonfinancial corporate sector DSR was rising sharply, and was approaching its early-2000 highs. During the pandemic, the corporate sector DSR has continued to rise in both regions, but this almost exclusively reflects a (temporary) decline in operating income, not a surge in corporate sector debt or a rise in interest rates. Not all of the pre-pandemic rise in the US corporate sector DSR was concentrated in broadly-defined tech stocks, but some of it likely was. The key point for investors is that the US nonfinancial corporate sector had a lower capacity to leverage itself relative to companies in the euro area at the onset of the pandemic, which implies a less accretive impact on relative earnings per share in the future. Chart II-13 reinforces this point by highlighting that the uptrend in relative cash flow for US growth stocks, versus global ex-US, appears to have ended in 2015. The uptrend has continued in per share terms, but this appears to be flattered by the impact of buybacks itself. Chart II-12Can The US Continue To Accrete EPS Through Stock Buybacks? Chart II-13US Growth Companies Are No Longer Generating More Cash Than Their Global Peers   Admittedly, we see no basis to conclude that the persistent earnings dilution that has occurred in emerging markets over the past 14 years will end, or even slow, over the secular horizon. This underscores that emerging markets will need to generate stronger revenue growth to prevent the dilution effect from acting as a continued drag on EM vs. US equity performance, and it is an open question as to whether this will occur. Thus, for now, we have more conviction in the view that capital structure and index composition changes may contribute less to US equity outperformance versus developed markets ex-US over the coming several years. Equity Multiples There are three arguments against the idea that US equity multiples will continue to expand relative to those of global ex-US stocks. First, Chart II-14 highlights a point that we have made in previous Bank Credit Analyst reports, which is that aggressive multiple expansion in the US has now rendered US stocks to be the most dependent on low long-maturity bond yields than at any point since the global financial crisis. Chart II-14US Stocks Are The Most Dependent On Low Bond Yields In Over A Decade Over the coming 6- to 12-months, we strongly doubt that US 10-year Treasury yields will rise outside of the range that would be consistent with the US equity risk premium from 2002 to 2007 (discussed in further detail in the next section). But the chart also shows that this range is now clearly below trend nominal GDP growth, suggesting that higher interest rates on a structural basis may cause outright multiple contraction for US stocks. This is particularly true for growth stocks, which have been responsible for a significant portion of US equity outperformance, given their comparatively long earnings duration. Chart II-15US Multiples Are Not Justified By Higher Return On Equity Second, it has been often argued by some investors that a premium is warranted for US stocks given their comparatively high return on equity, but Chart II-15 highlights that this is not the case. The chart shows the relative price-to-book ratio for the US versus global and developed markets ex-US compared with regression-based predicted values based on relative return on equity. The chart clearly highlights that the US price-to-book ratio is meaningfully higher than it should be relative to global stocks, especially when compared to other developed markets. Versus DM ex-US, the only comparable period that saw a relative P/B – relative ROE deviation of this magnitude occurred in the late-1980s, when US stocks were meaningfully less expensive than relative ROE would have suggested. This relationship completely normalized in the years that followed, which would imply a substantial relative multiple contraction for US stocks over the coming several years were the gap shown in Chart II-15 to close. Third, Chart II-16 presents the share of US stock market capitalization accounted for by the largest 10% of stocks by size. The chart highlights that the concentration of US market capitalization has risen to an extreme level that has only been reached in two other cases over the past century. Historically, prior stock market concentration has been associated with future increases in the equity risk premium, underscoring that broadly-defined US tech sector concentration bodes poorly for future returns. Chart II-16The US Stock Market Is Now Extremely Concentrated The Foreign Exchange Effect As a final point, Chart II-17 illustrates the degree to which US relative performance has meaningfully benefitted from a rise in the US dollar since 2008. The chart highlights that an equity market-weighted dollar index has risen 20% from its late-2007 level, which has boosted US common currency relative performance. The US dollar was arguably modestly undervalued just prior to the 2008/2009 global financial crisis, but Chart II-18 highlights that it is now meaningfully overvalued versus other major currencies. Over a multi-year horizon, this argues against further relative common currency gains for US stocks from the foreign exchange effect. Chart II-17The US Dollar Has Helped US Common Currency Performance... Chart II-18…And Is Now Expensive   The Absolute Secular Return Outlook For US Stocks Over a secular horizon, the most common method for forecasting equity returns is to predict whether earnings are likely to grow faster or slower than nominal potential GDP growth, and whether equity multiples are likely to rise or fall. For the reasons described above, we have no plausible basis on which to forecast that US profit margins are inclined to rise further over time given how extended they have become. This suggests that a reasonable long-term earnings forecast should be closely linked to one’s forecast for revenue growth. Chart II-19S&P 500 Revenue Is Low Relative To US GDP, And May Rise Over The Next Decade Chart II-19 presents S&P 500 revenue as a percent of nominal GDP, and underscores a fact that we noted above: revenue growth for US equities has underperformed US GDP since the global financial crisis. This undoubtedly has been linked to the fallout from the crisis and other exogenous shocks like the massive decline in energy prices in 2014/2015, which are unlikely to be repeated. Over the next ten years, the US Congressional Budget Office is forecasting nominal potential growth of roughly 4%; allowing for a potential rise in US equity revenue to GDP suggests that investors should expect earnings growth on the order of 4-5% per year over the coming decade, if extremely elevated profit margins are sustained. Chart II-20Multiples Seem To Predict Future Returns Well… Unfortunately for equity investors, there are slim odds that US equity multiples will continue to rise or even stay at their current level. Equity valuation has been shown to have nearly zero ability to predict stock returns over a 6-12 month time horizon or even over the following 3-5 years, but 10-year regressions relating current valuations on future 10-year compound returns tend to be highly predictive (Chart II-20). Utilizing this approach, today’s 12-month forward P/E ratio would imply a 10-year future total return of just 2.9% (Chart II-21). That, in turn, would imply a annual drag of 3-4% from multiple contraction over the coming decade, given our 4-5% earnings growth forecast and a historically average dividend yield of roughly 2%. One problem with the method shown in Charts II-20 and II-21 is the fact that the relationship between today’s P/E ratio and 10-year future returns captures more than the impact of potentially mean-reverting multiples. It also includes any correlation between the starting point of valuation and subsequent earnings growth, which is likely to be spurious. This effect turns out to be important: we can see in Chart II-21 that the strong fit of the relationship is influenced by the fact that the global financial crisis occurred roughly 10-years after the equity market bubble of the late-1990s. Chart II-21...But That Depends Heavily On The Tech Bubble / GFC Relationship Astute investors may infer a legitimate causal link between these two events, via too-easy monetary policy. But from the perspective of forecasting, predicting future returns based on prevailing equity multiples confusingly mixes together three effects: the relative timing of business cycles, the impact of changes in interest rates, and the potential mean-reverting nature of the equity risk premium. In order to disentangle these effects for the purposes of forecasting, we present a long-history estimate of the US equity risk premium based on Robert Shiller’s Irrational Exuberance dataset (Chart II-22). We define the equity risk premium as earnings per share (as reported) as a percent of the S&P 500, minus the real long-maturity interest rate. We calculate the real rate by subtracting the BCA adaptive inflation expectations model – essentially an exponentially smoothed version of actual inflation – from the nominal long-term bond yield. Chart II-22The US ERP Seems Normal Based On A Very Long Term History... The chart highlights that this estimate of the ERP is currently exactly in line with its median value since 1872. Chart II-23 presents essentially the same conclusion, based on data since 1979, using the forward operating P/E ratio for the S&P 500 and the same definition for real bond yields. This implies that, if interest rates were at equilibrium levels, investors would have a reasonable basis to conclude that equity multiples would be unchanged over a secular investment horizon. However, as we have highlighted several times in previous reports, long-maturity government bond yields are likely well below equilibrium levels. Chart II-24 highlights that long-maturity US government bond yields have not been this low relative to trend growth since the late-1970s. Chart II-23...And Based On The Forward Earnings Yield Over The Past Four Decades Chart II-24Interest Rates Are Well Below Equilibrium, And Are Likely To Rise Over Time   We presented in an April report why a gap between interest rates and trend rates of growth was indeed justified for a few years following the global financial crisis, but that a decline in the equilibrium real rate of interest (“r-star”) only appeared to be permanent due to persistent, non-monetary policy shocks to aggregate demand that occurred over the course of the last economic cycle.4 In a scenario where the US output gap turns positive, inflation rises modestly above target, and where permanent damage to the labor market from the pandemic is relatively limited over the coming 6-18 months, it seems reasonable to conclude that the narrative of secular stagnation may ultimately be challenged and that investor expectations for the neutral rate may converge toward trend rates of economic growth. This would weigh on equity multiples, and thus lower equity total returns from the 6-7% implied by our earnings forecast and income return assumption. Chart II-25US Stocks Are Likely To Earn Annual Total Returns Between 1.8-4.7% Over The Next Decade Were real long-maturity bond yields to rise by 100-200bps over the coming decade, this would imply annualized total returns of between 1.8 - 4.7% from US stocks, assuming 4-5% annual revenue growth, flat profit margins, a constant 2% dividend yield, and a constant ERP (Chart II-25). While this would beat the returns offered by bonds, implying that investors should still be structurally overweight equities versus fixed-income assets, it would also fall meaningfully short of the average pension fund return objective (Chart II-26), as well as the absolute return goals of many investors. Chart II-26Future Returns From US Stocks Will Greatly Disappoint Investors Investment Conclusions Chart II-27Over The Coming Year, Favor Value And Global Ex-US Stocks Over the coming 6-12 months, our view that 10-year US Treasury yields are likely to rise supports an overweight stance toward value versus growth stocks. Chart II-27 highlights that the underperformance of growth argues for an underweight stance toward US stocks within a global equity portfolio, especially versus developed markets ex-US. Over a longer-term horizon, there are two key investment implications from our research. First, the risks that we have highlighted to the sources of US outperformance over the past 14 years suggests that investors should not bank on a continuation of this trend over the next decade. We have not made the case in this report for the outperformance of global ex-US stocks, merely that the continued outperformance of US stocks now rests on an unreliable foundation. This may suggest that US relative performance will be flat over the structural horizon, arguing for a neutral strategic allocation. But even the cessation of US outperformance would be a significant development, as it would end the most consequential trend in regional equity performance in the post-GFC era. Second, investors should expect meaningfully lower absolute returns from US stocks over the next decade than what they have earned since 2008/2009, barring a continued rise in the already stretched profit margins of broadly-defined tech stocks. A structurally overweight stance is still warranted toward equities versus fixed-income, but even a 100% equity allocation is unlikely to meet investor return expectations in the high single-digits. As a consequence, global investors should be prepared to accept more volatility in order to reduce the gap between expected and desired returns, and should look towards riskier investments and asset classes (such as real estate and alternative investments) as potential portfolio return enhancements. Jonathan LaBerge, CFA Vice President The Bank Credit Analyst III. Indicators And Reference Charts BCA’s equity indicators highlight that the “easy” money from expectations of an eventual end to the pandemic have already been made. Our technical, valuation, and sentiment indicators remain very extended, highlighting that investors should expect positive but modest returns from stocks over the coming 6-12 months. Our monetary indicator has retreated below the boom/bust line, although this mostly reflects the use of producer prices to deflate money growth. In nominal terms, the supply of money continues to grow. Still, the retreat in the indicator highlights that the monetary policy stance is likely to shift in a tighter direction over the coming year. Forward equity earnings are pricing in a substantial further rise in earnings per share, and there is no meaningful sign of waning forward earnings momentum as net revisions and positive earnings surprises remain near record highs. Bottom-up analyst earning expectations are now almost certainly too high, but stocks are likely to be supported by robust revenue growth over the coming year. Within a global equity portfolio, global ex-US equities have underperformed alongside cyclical sectors, banks, and value stocks more generally. On a 12-month time horizon, we would recommend that investors position for the underperformance of financial assets that are negatively correlated with long-maturity government bond yields. The US 10-Year Treasury yield has broken above its 200-day moving average, beginning its recovery after falling sharply since mid-March. After a decline initially caused by waning growth momentum and the impact of the Delta variant of SARS-COV-2, long-maturity bond yields appear to be responding to the interest rate guidance that the Fed has been providing. 10-Year Treasury Yields remain below the fair value implied by a late-2022 rate hike scenario, underscoring that 10-Year Yields are set to trend higher over the coming year. The extreme rise in some commodity prices over the past several months has eased. Lumber prices have almost fully normalized, whereas the pace of advance in industrial metals prices has eased. Global shipping costs have exploded due to supply-side constraints, but are likely to ease over the coming year if further COVID-related labor market shocks can be avoided. US and global LEIs remain very elevated but have started to roll over. Our global LEI diffusion index has declined very significantly, but this likely reflects the outsized impact of a few emerging market countries (whose vaccination progress is still lagging). Still-strong leading and coincident indicators underscore that the global demand for goods is robust, and that output is below pre-pandemic levels in most economies because of very weak services spending. The latter will recover significantly at some point over the coming year, as social distancing and other pandemic control measures disappear. EQUITIES: Chart III-1US Equity Indicators Chart III-2Willingness To Pay For Risk Chart III-3US Equity Sentiment Indicators   Chart III-4US Stock Market Breadth Chart III-5US Stock Market Valuation Chart III-6US Earnings Chart III-7Global Stock Market And Earnings: Relative Performance Chart III-8Global Stock Market And Earnings: Relative Performance   FIXED INCOME: Chart III-9US Treasurys And Valuations Chart III-10Yield Curve Slopes Chart III-11Selected US Bond Yields Chart III-1210-Year Treasury Yield ComponentsChart III-13US Corporate Bonds And Health Monitor Chart III-14Global Bonds: Developed Markets Chart III-15Global Bonds: Emerging Markets   CURRENCIES: Chart III-16US Dollar And PPP Chart III-17US Dollar And Indicator Chart III-18US Dollar Fundamentals Chart III-19Japanese Yen Technicals Chart III-20Euro Technicals Chart III-21Euro/Yen Technicals Chart III-22Euro/Pound Technicals   COMMODITIES: Chart III-23Broad Commodity Indicators Chart III-24Commodity Prices Chart III-25Commodity Prices Chart III-26Commodity Sentiment Chart III-27Speculative Positioning   ECONOMY: Chart III-28US And Global Macro Backdrop Chart III-29US Macro Snapshot Chart III-30US Growth Outlook Chart III-31US Cyclical Spending Chart III-32US Labor Market Chart III-33US Consumption Chart III-34US Housing Chart III-35US Debt And Deleveraging   Chart III-36US Financial Conditions Chart III-37Global Economic Snapshot: Europe Chart III-38Global Economic Snapshot: China   Jonathan LaBerge, CFA Vice President The Bank Credit Analyst Footnotes 1  Please see The Bank Credit Analyst "The Return To Maximum Employment: It May Be Faster Than You Think," dated August 26, 2021, available at bca.bcaresearch.com 2  Please see China Investment Strategy "A Quick Take On Embattled Evergrande," dated September 21, 2021, and China Investment Strategy "The Evergrande Saga Continues," dated September 29, 2021, available at cis.bcaresearch.com 3 Please see US Political Strategy "Forget Biden's Budget," dated June 2, 2021, available at usps.bcaresearch.com 4     Please see The Bank Credit Analyst “R-star, And The Structural Risk To Stocks,” dated March 31, 2021, available at bca.bcaresearch.com

Dear Client, Next week, in lieu of our regular weekly report, I will be hosting two webcasts where I will discuss our view on China’s economy and financial markets. I will also address the topics that our clients are most concerned about, including China’s regulatory developments, inflation, and policy direction. The webcasts will be held on Wednesday, September 22 at 10:00 am EDT (English), and Thursday, September 23 at 9:00 am HKT (Mandarin). I look forward to discussing with you during the webcast. We will return to our regular publishing schedule on Wednesday, September 29. Best regards, Jing Sima, China Strategist Highlights China is facing cyclical inflationary pressures more than disinflationary ones. Prices of mining, raw materials and manufacturing goods have been rising at record rates. Chinese manufacturers are operating at close to full production, which suggests that there is little slack in demand. Despite soft headline readings in consumer prices, the costs of goods and services have rebounded to pre-pandemic levels. Prices for home durable goods, fuel and utilities have surged to multiyear highs. Measures to boost domestic demand will be limited as long as inflationary pressures continue and manufacturers produce at close to full capacity. Near-term policy support will likely focus on reducing costs for manufacturers and improving wage growth for lower-income households. We are initiating a trade: long industrial stocks/short A-shares.  Feature China’s Producer Price index (PPI) registered a 13-year high in August, at the time when the domestic economy continued to slow. On the other hand, consumer prices (CPI) - both headline and core CPI - have been lackluster. The acceleration in producer inflation and the demand dynamics raise the question whether China is in a stagflation, a situation in which prices climb but wages and demand do not follow. Consequentially, economy policy faces a dilemma between boosting demand and containing inflation. Inflationary pressures have been driven by pandemic-related factors and the supply-side constraints will likely continue into Q1 next year. These inflationary pressures, and more importantly, undercurrents in the inflation prints, will constrain Chinese policymakers’ efforts to reflate the economy. The recent rebound in Chinese infrastructure stocks is overdone. Material stocks are also vulnerable to price setbacks. Global commodity prices will soften, although from very elevated levels. Meanwhile, we are initiating a trade: long Chinese industrial stocks relative to the A-share market. Despite falling profit growth in recent months, China’s leadership is increasing its support, both cyclically and structurally, to the manufacturing sector. Inflation Or Deflation? The details in both the PPI and CPI readings indicate that China is facing more inflationary pressures than disinflationary ones. Producers are raising prices across the board. Although consumer prices will likely remain well below the PBoC's 3% inflation target for the year mainly due to low food prices, prices in some of the key consumer goods segments are rising at an alarming pace. The inflationary pressures will continue for producers, at least through the first quarter of 2022. The strength in August’s PPI was concentrated in mining and raw materials (Chart 1, top panel). Robust global demand and tight supply conditions supported high oil and base metals prices, while pushing up coal prices. Chart 1Chinese Mining And Manufacturing Goods Prices Accelerated To Record Highs Chart 2Commodity Prices Held Up Despite A Slowing China We do not expect China’s infrastructure investment growth to pick up and support industrial metal prices. However, this year’s unsynchronized recovery in global demand and severe supply shortages have delayed the global commodity market’s price reaction to slowing Chinese demand (Chart 2). Moreover, as China’s environmental policy remains stringent during the upcoming winter, supply-side constraints from production cuts will partially offset the slowdown in China’s demand for mining and raw materials (Chart 3A and 3B). Chart 3ASupply-Side Constraints And Chinese Production Cuts Likely To Continue Into Early 2022 Chart 3BSupply-Side Constraints And Chinese Production Cuts Likely To Continue Into Early 2022 Manufacturing goods inflation registered its topmost annual growth since data collection started in 1996 (Chart 1, bottom panel). Moreover, capacity utilization rates in the industrial and manufacturing sectors are at the highest levels since 2007, well above their means (Chart 4). Changes in manufacturing capacity are highly correlated with China’s export growth and tightly linked to PPI (Chart 5). Therefore, manufacturing goods prices will remain lofty as long as external demand stays robust and China’s manufacturers continue to produce near maximum output. Chart 4Chinese Manufacturers Are Producing Near Their Max Capacity Chart 5Robust Exports Have Been Supporting Strong Chinese Manufacturing Output The PPI’s weakest component has been consumer goods, which inched up by a mere 0.3% from a year ago (Chart 6). However, consumer goods only account for 25% of PPI, whereas industrial and manufacturing producer goods are 75%. In addition, the underlying data shows that among the four sub-components in the PPI’s consumer goods, only food prices have remained below their pre-pandemic levels (Chart 7, top panel). Prices in durable goods have rebounded strongly since March last year and clothing and daily sundry articles have recovered to their end-2019 rate of growth (Chart 7, mid and bottom panels).  Chart 6Producer Prices For Consumer Goods Remain Soft... Chart 7...But Food Prices Have Been The Main Drag The PPI’s price forces are consistent with the CPI, in which food has been the main drag. Core CPI, along with prices for consumer goods and services, have returned to pre-pandemic growth rates (Chart 8). Durable goods prices, such as home appliances, increased to a multiyear high in August. Fuel and utilities costs have also risen. This suggests that despite the soft CPI readings, inflation has flowed from producers to Chinese consumers through manufacturing goods. The passthrough will likely intensify into Q4 when domestic COVID-cases have been largely brought under control and the September – October holiday season will boost consumption for both goods and services. Chart 8Prices For Other Consumer Goods Categories Have Recovered Table 1A Look At China’s CPI Basket – Food Dominates We still expect that headline CPI will remain below the PBoC’s 3% inflation target for the year. Consumer durable goods prices are lightly weighted in China’s CPI, therefore, an acceleration in inflation passthroughs in this component is unlikely to significantly push up the CPI aggregates (Table 1). Chart 9Prices For Healthcare And Education Services On A Structural Downshift In addition, there are some structural headwinds that will affect prices in the education and healthcare and medical services components, which together account for about 15% of the CPI. Healthcare prices have been on a policy-driven structural downshift since late 2017 and recent regulatory changes in the education industry will depress pricing power in that sector (Chart 9). Despite sluggish aggregate consumer prices, climbing prices in consumer durable goods, services and particularly, fuel and utilities, will likely force China’s leadership to take action on policy. Bottom Line: Price pressures for Chinese producers remain intense and consumers will feel the heat of escalating prices in durable goods, fuel and utilities. Inflation is threatening domestic demand, which is already slowing from its peak earlier this year. Implications On Policy Response Inflation readings –even though they are lagging economic indicators –bear significant forward-looking market implications because changes in inflation dynamics herald various policy responses. Despite slower economic growth, higher inflation coupled with accommodative monetary and fiscal policies may indicate that the economy is in a “goldilocks” stage and corporate profits can still benefit (Chart 10). Chinese onshore stocks reached record high recently (Chart 11). Chart 10Are Chinese Corporates In A 'Sweet Spot'? Chart 11Accommodative Monetary Conditions Propelled Chinese Stock Prices To Highest Since 2015 However, underlying trends in China’s producer and consumer inflation prints raise the risks that policymakers may not deliver the ingredients needed for a “just right” scenario. Even though China has kept a loose monetary policy that we expect to extend into next year, inflationary pressures may force policymakers to either delay or reduce the magnitude of stimulus. Recent policy moves show that the authorities are focused on reducing input cost burdens and bumping up support for small- and medium-sized enterprises (SMEs), which are highly concentrated in mid- to downstream manufacturing and services sectors. In our view, the recent rhetoric from policymakers further reduces the odds of any broadly based stimulus to boost demand. Our view is based on the following observations: The elevated global input costs and limited price passthroughs to consumers are depressing Chinese manufacturers’ profit margins and incentives to expand production capacity. Despite strong exports and production, manufacturing investment has lagged that in infrastructure and real estate this year (Chart 12). Consumers, particularly lower-income households, are bearing most of the burdens; rising costs and slow wage growth are weakening their propensity to spend (Chart 13). Chart 12Slower Manufacturing Investment Recovery Than Infrastructure And Real Estate So Far This Year Chart 13Slow Wage Growth Limits The Pace Of Consumption Recovery The inflation prints came at the time when China’s top leadership shifted its structural policy goals to reduce income inequality and stabilize manufacturing share in the aggregate economy. The structural goals will likely be reflected in policy responses to the cyclical challenge.  Moreover, this year’s manufacturing production volume was growing twice as fast as producer prices, a reversal from 2017 when price increases outpaced production (Chart 14). Price changes are much more important to corporate profits than volume changes. A strong RMB and sharply escalating shipping costs have also reduced exporters’ pricing power and profits (Chart 15). In contrast, mounting prices across various commodities have allowed the upstream industrial sectors, which are dominated by SOEs, to deliver much stronger profits than the downstream and private sector (Chart 16). Chart 14Growth In Manufacturing Output And Prices Starting To Converge Chart 15Strong RMB And Rising Shipping Costs Have Reduced Chinese Exporters' Profitability   It is unsurprising that authorities are increasing support to the private sector in order to maintain manufacturing share in the economy and keep the export sector competitive (Chart 17). A boost in infrastructure investment, on the other hand, would exacerbate upward pressure on commodity prices and mostly benefit upstream SOEs. Chart 16Upstream Industries Disproportionally Benefited From Surging Commodity Prices Chart 17Private Sector: Lower Profit Margin, Higher Costs Furthermore, stimulating the traditional sectors would not revive household consumption. The subdued recovery in consumption and prices for consumer staple goods is due to slow growth in lower-income household wages and a disrupted recovery in the services sector. Ramping up infrastructure investment can support headline GDP growth, but will do little to provide jobs and wages since China’s private sector provides 80% of all jobs and 90% of annual job creations. Lower-income households have a higher marginal propensity to consume. We expect the government to accelerate fiscal support measures to fortify wages among lower-income households. Bottom Line: Ongoing inflationary pressures and the underlying forces will likely thwart policymakers from stepping up their efforts to stimulate the old economy sectors. Investment Conclusions Chart 18Rebound In Infrastructure Stocks Should Be Short-Lived Chinese onshore stocks in the infrastructure, materials, and industrial sectors recently advanced strongly in the expectation that policymakers will ramp up their fiscal support in the old economy sectors, particularly infrastructure. Although we agree that infrastructure investment will improve, we maintain our view that a sizable rebound is highly unlikely this year. Hence, we do not expect that the rally in infrastructure stocks will be long-lasting (Chart 18).  We are probably too late in the cycle to re-initiate our long material/broad market trade in the onshore and offshore equity markets (Chart 19). We closed the trade in December last year when Chinese policymakers started pulling back stimulus, and in expectations that raw material prices would tumble. However, we underestimated the intensity of China’s de-carbonization efforts and protracted global supply-side constraints. Although global commodity prices will remain elevated into 2022, the price rallies from this year are not sustainable on a cyclical (6- to 12-month) basis. Therefore, we do not recommend material stocks as a cyclical play.  Chart 19Price Rally In Materials Stocks Unlikely To Sustain Chart 20Industrial Stocks May Be On A Structural Upcycle Instead, we recommend a long industrial/broad A-share market trade (Chart 20). Even though China is in a late business cycle and the upcoming stimulus will be mediocre at best, we think that the industrial sector will benefit from policy support for investment in the manufacturing sector and a faster pace in the sector’s capacity expansion.   Jing Sima China Strategist jings@bcaresearch.com   Footnotes Market/Sector Recommendations Cyclical Investment Stance

Special Report

Highlights Germany’s election on September 26 is more of an opportunity than a risk for global investors. Coalition formation will prolong uncertainty but the key takeaway is that early or aggressive fiscal tightening is off the table for Germany … and hence the EU. Germany’s left wing is surprising to the upside as predicted, but it is the Social Democrats rather than the Greens who have momentum in the polls. This is a market-positive development. A coalition of only left-wing parties is entirely possible, but there is a 65% chance that the Christian Democrats (or Free Democrats) will take part in the next coalition to get a majority government. This would constrain business unfriendly outcomes. The German economy is likely to slow for the remainder of 2021, but the outlook for 2022 remains bright as the current headwinds facing the country will dissipate, especially if the risk of an aggressive fiscal drag is low. The underperformance of German equities relative to their Eurozone counterparts is long in the tooth. A combination of valuation, earnings momentum and technical factors suggests that German stocks will beat their peers next year. German equities will also outperform Bunds, which offer particularly unattractive prospective returns. Feature Germany’s federal election will be held on September 26. Our forecast that the left wing will surprise to the upside remains on track, albeit with the Social Democrats rather than the Greens surging to the forefront of opinion polls (Chart 1). However, the precise composition of the next government is very much in the air. Chart 1German Election: Social Democrats Take The Lead Our quantitative German election model – which we introduce in this special report – predicts that the ruling Christian Democratic Union will outperform their current 21% standing in opinion polls, winning as much as 33% of the popular vote. Subjectively, this seems like an overestimation, but it goes to show that outgoing Chancellor Angela Merkel’s popularity, a historically strong voting base, and the economic recovery will help the party pare its losses this year. This finding, combined with the strong momentum for the Social Democrats, suggests that the election outcome will not be decisive. Germany will end up with either a grand coalition that includes Merkel’s Christian Democrats or a left-wing coalition that lacks a majority in parliament.1 Investors should note that none of the election outcomes are hugely disruptive to domestic or foreign policy. The status quo is unexciting but not market-negative, while a surprise left-wing victory would mean more reflation in the short run but a roll back of some pro-business policies in the long run. More broadly Germany has established a national consensus that rests on European integration, looser fiscal policy, renewable energy, and qualified engagement with autocratic powers like Russia and China. The chief takeaway is that fiscal policy will not be tightened too soon – and could be loosened substantially. Germany’s Fiscal Question Outgoing Chancellor Angela Merkel is stepping down after ruling Germany since 2005. The Christian Democratic Union, and its Bavarian sister party the Christian Social Union, together form the “Union” that is hard to beat in German elections, having occupied the chancellor’s office for 57 out of 72 years. However, both the Christian Democrats and the Social Democrats, their main rivals, have been shedding popular vote share since 1990, as other parties like the Greens, Free Democrats, the Left, and Alternative for Germany have gained traction (Table 1). Table 1Germany: Traditional Parties Lose Vote Share Over Time The Great Recession and European sovereign debt crisis ushered in a new geopolitical and macroeconomic context that Merkel reluctantly helped Germany and the EU navigate. Germany’s clashes with the European periphery ultimately resulted in deeper EU integration, in accordance with Germany’s grand strategy and Merkel’s own strategy. But just as the euro crisis receded, a series of shocks elsewhere threatened to upend Germany’s position as one of the biggest economic winners of the post-Cold War world. The sluggish aftermath of the financial crisis, the Russian invasion of Crimea, the Syrian refugee crisis, the Brexit referendum, and President Trump’s election in the US sparked a retreat from globalization, a direct threat to an export-oriented manufacturing economy like Germany. In the 2017 election the Union lost 13.4 percentage points compared to the 2013 election. Minor parties have gradually gained ground since then. However, through a coalition with the Social Democrats, Merkel and her party managed to retain control of the government. This grand coalition eased the country’s fiscal belt in response to the trade war and global slowdown in 2019, signaling Germany’s own shift away from fiscal austerity. Then COVID-19 struck, prompting a much larger fiscal expansion to tide over the economy amid social lockdowns. Germany was not the largest EU member in terms of fiscal stimulus but nor was it the smallest (Chart 2). It joined with France to negotiate a mutual debt plan to rescue the broader EU economy and deepen integration. Chart 2Germany’s Fiscal Stimulus Ranks In The Middle Of Major Countries Germany’s pro-EU perspective has been reinforced by Brexit and is not on the ballot in 2021. Immigration and terrorism have temporarily subsided as voter concerns. The focus of the 2021 election is how to get through the pandemic and rebuild the German economy for the future. For investors the chief question is whether conservatives will have enough sway in the next government to try to semi-normalize policy and consolidate budgets in the coming years, or whether a left-wing coalition will take charge, expanding on Germany’s proactive fiscal turn. The latter has consequences for broader EU fiscal normalization as well since Germany is traditionally the prime enforcer of deficit limits. The latest opinion polls point to more proactive fiscal policy. The country’s left-leaning ideological bloc has taken the lead (Chart 3A) and the Social Democratic leader Olaf Scholz has sprung into first place among the chancellor candidates (Chart 3B). Chart 3AGermany: Voting Intentions Favor Left-Leaning Parties Chart 3BSocial Democrats Likely To Take Chancellery Scholz has served as finance minister and is the face of the country’s recent fiscal stimulus efforts. Public opinion is clearly rewarding him for this stance as well as his party, which was previously in the doldrums.2 The Social Democrats and Greens are calling for more fiscal expansion as well as wage hikes and tax hikes (wealth redistribution) in pursuit of social equality and a greener economy (Table 2). If the Christian Democrats retain a significant role in the future coalition, these initiatives will be blunted – not to say halted entirely. But if the left parties put together a ruling coalition without the Christian Democrats, then they will be able to launch more ambitious tax-and-spend policies. Opinion polls show that voters still slightly favor coalitions that include the Christian Democrats, although momentum has shifted sharply in favor of a left-wing coalition (Chart 4). Table 2German Party Platforms Chart 4Voters Evenly Split On Whether Next Coalition Should Include CDU This shift is what we forecast in previous reports but now the question is whether the left-wing parties can actually win enough seats to put together a majority coalition. That is a tall order. Our quantitative election model suggests that the Christian Democrats, having suffered a long overdue downgrade in expectations, will not utterly collapse when the final vote is tallied. While we do not expect them to retain the chancellorship, momentum will have to shift even further in the opposition’s favor over the next two weeks to produce a majority coalition that excludes the Union. Our Quantitative German Election Model Our model is based off the work of Norpoth and Geschwend, who created a simple linear model to predict the vote share that incumbent governing parties or coalitions will obtain in impending elections.3 Their model utilizes three explanatory variables and has a sample size of 18 previous elections, covering elections from 1953 to 2017. Our model updates their original work to make estimates for the 2021 election. Unlike our US Political Strategy Presidential Model, which makes use of both political and economic explanatory variables in real time, our German election model makes predictions based solely on historical political variables, all of which display a high degree of correlation with popular vote share. We will look at economic factors that may affect the election later in this report. The Three Explanatory Variables 1. Chancellor Approval Rating: This variable captures the short-term support rate of the incumbent chancellor. A positive relationship exists between chancellor approval and vote share: higher approval equates to higher vote share for the incumbent party. Merkel’s approval stands at 64% today which is a boon for the otherwise beleaguered Christian Democrats (Chart 5). Chart 5Merkel's Coattails A Boon But Not Enough To Save Her Party 2. Long-term partisanship: This variable shows the long-term support rate of voters for specific parties or coalitions in past elections. It is measured as the average vote share of the incumbent party over the past three elections. A positive relationship with vote share exists here too: higher historical partisanship equates to a higher share of votes in forthcoming elections, and vice versa. This variable clearly gives a boost to the Christian Democrats – although it could overrate them based on past performance, as occurred in 2017 when they underperformed the model’s prediction.4 3. “Time For Change”: This is a categorical variable measured by how many terms the parties or coalition have held office leading into an election. This variable has a negative relationship with vote share outcomes. The longer an incumbent party or coalition holds office, the less vote share they will receive. Effectively, our model punishes parties that hold office for long periods of time. In this case that would be the long-ruling Christian Democrats. Model Estimation And Results Our model is estimated by the following simple equation: Popular Vote Share = constant + ßChancellor Approval Rating + ßLong-Term Partisanship + ßTime For Change Estimating the above model for the 2021 election predicts that the Union will win 32.7% of the vote share (Table 3). If this prediction came true, it would suggest that the ruling party performed almost exactly the same as in 2017. In other words, the party’s strong voter base combined with Merkel’s long coattails are expected to shore up the party. This flies in opinion polling, however, so we think the model is overestimating the Christian Democrats. Table 3Our German Election Quant Model Says CDU Will Not Collapse Note that even if the Union performs this well, it still will not win enough seats to govern on its own. Potential Union-led coalitions are shown in Table 3, excluding the Social Democrats (see below). For a majority government, a coalition with the Free Democrats and the Greens would need to be formed. This coalition would equate to 53% of the vote share. Otherwise, to obtain a majority, the Union would have to team up with the Social Democrats, which is today’s status quo. We can use the same methodology to predict the vote share for the Social Democrats. We use the support rate of Social Democratic chancellor-candidate Olaf Scholz and calculate the long-term partisanship variable using past Social Democratic vote shares. In this case our model predicts that the Social Democrats will win 22.1% of the vote. If this result were to come true, it would not be enough for the party to govern own its own. Potential Social Democratic-led coalitions are shown in Table 4. The best coalition would be with the Greens and either the Left or the Free Democrats. But in this case the Social Democrats cannot form a government with a vote share above 50%, unless it pairs up with the Christian Democrats. Table 4Our German Election Quant Model Says SPD Has Not Yet Won It All In other words, either the left-wing parties must build on their current momentum and outperform their historical record in the final election tally, or they will need to form a coalition with the Christian Democrats. This kind of left-wing surge is precisely what we have predicted. But the model helps put into perspective how difficult it will be for the left-leaning parties to get a majority. Scholz is single-handedly trying to overcome the long downtrend of the Social Democrats. His party is rising at the expense of the Greens, and the Left, which puts a lid on the total left-wing coalition size. If these three parties all beat the model and slightly surpass their top vote share in recent memory (SPD at 26%, Greens at 11%, and the Left at 12%), they still only have 49% of the vote. While our model is reliant on historical political data, it is a robust predictor for past election results (Chart 6). The average vote share error between the predicted and realized outcomes over from 1953 to 2013 is 1.7 percentage points. The problem with relying on the model is that the Christian Democrats have broken down from their long-term trend in opinion polls. And while Merkel’s approval is strong, she is no longer on the ballot and her hand-picked successor, Armin Laschet, is floundering in the polls (see Chart 3B above). Chart 6Our German Election Quant Model Has Solid Track Record, But Merkel’s High Approval Rating Caused Overestimate In 2017 And May Do So In 2021 In short, the model is probably overrating the Union but it is also calling attention to the extreme difficulty of the left-wing parties forming a majority coalition. Scholz may have to form a coalition with the Free Democrats or pursue another grand coalition. And if the Social Democrats fail to get the largest vote share, German President Frank-Walter Steinmeier may ask Armin Laschet to try to form a government first. Still, Scholz is the most likely chancellor when all is said and done. Election Model Takeaway Our German election model predicts that the Union will receive 32.9% of the popular vote, while the Social Democrats will receive 22.1%. At the same time, the left-leaning parties, specifically the Social Democrats, clearly have the momentum. Therefore the model may be overrating the incumbent party. But it still calls attention to a high level of uncertainty, the likelihood of a messy election outcome, and a tricky period of coalition formation. The Social Democrats will have to pull off a major surprise, outperforming both history and our model, to lead a majority government without the Christian Democrats.5 We still think this is possible. But we will stick with our earlier subjective probabilities: 65% odds that the Christian Democrats take part in the next coalition, 35% odds that they do not. Bottom Line: The chancellorship will go to the Social Democrats but the coalition will constrain the business unfriendly aspects of their agenda. This is positive for Germany’s corporate earnings outlook. Macro Outlook: A Temporary Economic Dip Our election model does not account for the economic backdrop and hence ignores the “pocketbook voter.” Germany is recovering from the pandemic, which is marginally supportive for an otherwise faltering ruling party. However, the economic data is only good enough to suggest that the Union will not utterly collapse. A rise in unemployment, inflation, and the combination of the two (the “Misery Index”) is a tell-tale sign that the incumbent party will suffer a substantial defeat (Chart 7). However the German economy’s loss of momentum is temporary. Growth will re-accelerate in early 2022. The timing is politically inconvenient for the ruling party but positive news for investors. German economic confidence is deteriorating. The Ifo Business Climate survey has rolled over, lowered by a meaningful decline in the Expectations Survey. Additionally, consumer confidence is turning south, despite already being low (Chart 8). Chart 7Spike In German Misery Index A Tell-Tale Sign Of Poor Election For Incumbent Party Chart 8Deteriorating German Confidence A combination of factors weighs on German confidence: First, global supply chain bottlenecks are hurting growth. The automotive industry, which is paralyzed by a global chip shortage, accounts for about 20% of industrial production, and its output is once again declining after a sharp but short-lived rebound last year (Chart 9). Similarly, inventories of finished goods are collapsing, which is hurting growth today (Chart 9, second panel). Second, the Delta variant of COVID-19 is causing a spike in infections. The rise in cases prevents containment measures from easing as much as expected, while it also hurts the willingness of households to go out and spend their funds (Chart 9, third panel). Third, German real wages are weak. Negotiated wages are only growing at a 1.7% annual rate, and wages and salaries are expanding at 2.1% annually. Meanwhile, German headline CPI runs at 3.9%. The declining purchasing power of German households accentuates their current malaise. Three crucial forces counterbalance these negatives: First, German house prices are growing at a 9.4% annual rate, which is creating a potent, positive wealth effect (Chart 10). Chart 9Germany's Headwinds Chart 10A Strong Wealth Effect Second, German household credit remains robust. According to the Bundesbank, the strength in household credit mostly reflects the strong demand for mortgages. Historically, a healthy housing sector is an excellent leading indicator of economic vigor. Third, the Chinese credit impulse is too depressed for Beijing’s political security. The recent decline in the credit impulse to -2.4% of GDP reflects a policy decision in the fall of 2020 to trim down the credit expansion. As a result, Chinese economic growth is slowing. For example, both the Caixin Manufacturing and Services PMIs stand below 50, at post-pandemic lows of 49.2 and 46.7, respectively. In July authorities became uncomfortable and cut the Reserve Requirement Ratio as well as interbank rates to free liquidity and stabilize the economy. A boom is not forthcoming, but the drag on global activity will ebb by next year. Including the headwinds and tailwinds to the economy, German activity will slow down for the remainder of the year before improving anew in 2022. Our election case outlined above – that the conservatives will lose the chancellorship and either be excluded from power or greatly diminished in the Bundestag – means that fiscal policy will not be tightened abruptly and will not create a material risk to this outlook. Chart 11Vaccines Work Many of the headwinds will dissipate. The Delta-wave of COVID-19 will diminish. Already, Germany’s R0 is tentatively peaking, which normally precedes a drop in daily new cases. Moreover, Germany’s vaccination campaign is progressing, which limits the impact of the current wave on hospitalization and intensive care-unit usage (Chart 11). Inflation will peak in Germany, which will salvage real wages. As European Investment Strategy wrote last Monday,6 European inflation remains concentrated in sectors linked to commodity prices or directly affected by bottlenecks. Instead, trimmed-mean CPI is muted (Chart 12), which implies that underlying inflationary pressures are small, especially as wage gains are still well contained. Moreover, the one-off impact of the end of the German VAT rebate will also pass. Finally, a stabilization and eventual revival of the Chinese credit impulse will put a floor under German exports, industrial production, and capex (Chart 13). For now, the previous decline in the Chinese credit impulse is consistent with slower German output growth for the remainder of 2021. However, next year, the German industrial sector will start to feel the effect of the current efforts to improve Chinese liquidity conditions. Chart 12Narrow European Inflation Bottom Line: The German economy is set to deteriorate for the remainder of 2021. However, as the current wave of COVID-19 infections ebbs, real wages recover, and China’s credit impulse stabilizes, Germany’s economic activity will re-accelerate in 2022, especially if the upcoming election does not generate a meaningful fiscal shock. We do not think it will. Chart 13China: From Headwinds To Tailwind? Market Implications: German Stocks To Shine German equities are set to outperform their European counterparts and will significantly beat Bunds over the coming 18 months. During the past 5 months, the German MSCI index has underperformed the rest of the Eurozone by 6.2%. The poor performance of German equities is worse than meets the eye. If we adjust for sectoral differences by building equal sector-weight indexes, Germany has underperformed the Euro Area by 22% since early 2017 (Chart 14). Chart 14Not Delivering The Goods This underperformance is long in the tooth and should reverse because of four important dynamics. First, German equities are cheap relative to the European benchmark. As Chart 15 highlights, the relative performance of German stock prices has lagged that of profits. This underperformance is also true once we account for the different sectoral composition of the German market. As a result, Germany is cheap on a forward price-to-earnings, price-to-sales, and price-to-book basis versus the Euro Area. Additionally, analysts embed significantly lower long-term and one-year expected growth rates of earnings in Germany than in the rest of the Eurozone, which depresses the German PEG ratios. Second, German operating metrics do not justify the valuation discount of German equities. The return on equity of German stocks stands at 11.39%, which is similar to that of the Euro Area. Profit margins are also comparable, at 5.91% and 5.74%, respectively. However, German firms utilize their capital more efficiently, and their asset turnover stands at 0.3 times compared to 0.2 times for the Eurozone average. Meanwhile, German non-financial firms are less indebted than their Eurozone competitors, which implies that Germany’s return on assets is greater than that of Europe at large (Chart 16). Chart 15Lagging Prices, Not Earnings Chart 16Why The Discount? Third, the drivers of earnings support a German outperformance. Over the past thirty years, commodity prices led the performance of German stocks relative to that of the rest of the Eurozone (Chart 17). While the near-term outlook for natural resource prices is muddy, BCA’s commodity strategists expect Brent prices to average more than $80/bbl in 2023 and industrial metals to outperform energy over the coming years.7 Additionally, German Services PMI are bottoming compared to that of the Eurozone. Over the past decade, this process preceded periods of outperformance by German stocks (Chart 18). Similarly, the collapse in the Chinese credit impulse relative to the robust domestic economic activity in Europe is well reflected in the underperformance of German shares. The Eurozone’s Service PMI is near all-time highs and unlikely to improve further; however, the Chinese credit impulse should recover in the coming quarters. This phenomenon will help German stocks (Chart 19). Chart 17Commodity Bulls Pull Germany Chart 18German Vs European Activity Matters Chart 19German Vs Chinese Activity Matters The German MSCI index is also oversold. The 52-week rate of change of its performance compared to the rest of the Eurozone plunged to its lowest reading since the introduction of the euro in 1999 (Chart 20). Meanwhile, the 13-week rate of change remains low but has begun to improve (not shown). This combination usually heralds a forthcoming rebound in German relative performance. In relation to equities, German Bunds remain an unappealing investment. Based on historical experience, the current yield of -0.36% offered by German 10-year bonds condemns investors to negative returns over the next five years (Chart 21). Chart 20Oversold! Chart 21Bounded Bunds' Returns Even if realized inflation ebbs in Germany and Europe, inflation expectations remain low and an eventual return to full employment will force CPI swaps higher, especially if the ECB maintains easy monetary conditions and invites further risk-taking in the Eurozone. The global economic cycle will also move from a friend to a foe for Bunds. As Chart 22 illustrates, the recent deceleration in global export growth was consistent with the fresh uptick in the returns of German paper. However, if Chinese credit flows stabilize by year-end and reaccelerate in 2022 while supply-chain bottlenecks dissipate, global export growth will improve. This should hurt Bund prices, especially as the long-term terminal rate proxy embedded in the German curve remains too low. As a result, not only should Bunds underperform German equities, but the German yield curve will also steepen further relative to that of the US, where the Fed will lift the short-end of the curve faster than the ECB. Chart 22Economic Momentum And Bunds Prices Bottom Line: The underperformance of German equities relative to those of the rest of the Eurozone is well advanced, which makes German stocks a bargain. The current deceleration in global and German growth will not extend beyond 2021, which suggests that German stocks prices should converge toward their earnings outperformance next year. Our political forecast suggests that the odds of an early or aggressive fiscal retrenchment are very low. Additionally, German equities will outperform Bunds, which offer particularly poor prospective returns.   Matt Gertken Vice President Geopolitical Strategy mattg@bcaresearch.com Mathieu Savary Senior Vice President Mathieu@bcaresearch.com Guy Russell Research Analyst GuyR@bcaresearch.com Jingnan Liu Research Associate JingnanL@bcaresearch.com Footnotes 1 Note that minority governments are rare and have a bad reputation in Germany, partly as a result of the series of weak governments leading up to the 1932 election and Nazi rule. 2 In addition, while the center-left parties can work with the far-left in the Bundestag, the center-right parties cannot work with the far-right Alternative for Germany. Indeed the slightest imputation of a willingness to work with Alternative for Germany cost Merkel’s first pick for successor, Annegret Kramp-Karrenbauer, her job. 3 See: Norpoth, Helmut & Gschwend, Thomas (2010) The chancellor model: Forecasting German elections, International Journal of Forecasting. 26. 42-53. 4 Our model performs well in back-testing but 2017 was an outlier. It correctly predicted the Union to win the highest share of the popular vote but overestimated that vote by seven percentage points. Our only short-term variable, the chancellor’s approval rate, caused a deviation from long-term voting trends. Our other two variables capture medium and long-term effects, which clearly favored the Union. The implication is that Merkel’s high approval rating today could give a misleading impression about the Christian Democrats’ prospects. 5 If they are forced to rely on the Free Democrats instead, that will also constrain the most anti-business elements of their agenda. 6 Please see BCA Research European Investment Strategy Weekly Report, "The ECB Taper Dilemma", dated September 6, 2021, available at eis.bcareseach.com. 7 Please see BCA Research Commodity & Energy Strategy Weekly Report, "Permian Output Approaches Pre-Covid Peak", dated August 19, 2021, available at ces.bcareseach.com.

Feature Chart 1Chinese Offshore Stocks Tumbled Amid Regulatory Crackdowns Relative to the global equity index, onshore and offshore Chinese stocks have fallen by 18% and 32%, respectively, since their peaks in mid-February (Chart 1). The panic sell-off in the offshore market, which saw greater losses due to its high concentration in internet stocks, appears to be overdone and may technically rebound in the near term. However, any short-term bounce in Chinese stocks from oversold levels will likely be short-lived (Chart 2). The crackdown on new economy companies reflects socio-political and economic shifts in China, which raises the odds that the restrictions will continue with further actions focused on social welfare and healthcare. August’s official PMIs and economic data indicate a broad-based softening in China’s domestic demand and production. However, compared with 2018/19 when the US-China trade war exacerbated the deterioration in an already slowing economy, the economy now remains well supported by strong exports. Moreover, the magnitude of the slowdown has not exceeded policymakers’ pain thresholds (Chart 3). Chart 2Tactical Bounce Was Short-Lived In Previous Downturns Chart 3China's Economic Recovery Losing Steam, But From An Elevated Level In 2018/19, stimulus was measured and the authorities did not meaningfully relax limits on bank lending standards and shadow banking. Furthermore, China recently reiterated its cross-cycle macro policy setting, which means that policymakers will not use significant stimulus to achieve high and short-term economic growth. Given financial stability measures that aim to contain risks associated with the housing market and hidden local government debt, any monetary and fiscal easing will likely help to stabilize credit growth instead of substantially boosting it this year. For the time being, China’s financial assets continue to face downside risks stemming from a confluence of a weakening business cycle and ongoing regulatory tightening. Thus, we recommend investors maintain an underweight allocation to Chinese equities within a global equity portfolio. Qingyun Xu, CFA Associate Editor qingyunx@bcaresearch.com     A Shining Moment For Chinese Small And Medium Caps Small and medium-cap (SMID-cap) stocks have outperformed large-caps since February and the recent regulatory restrictions have intensified the situation. The CSI500 index, which comprises 500 SMID-cap companies, has outperformed the large-cap CSI300 by 34% since mid-February (Chart 4, top panel). Uncertainties surrounding the pandemic and corporate earnings growth have fueled extreme dislocations between large-cap and SMID-cap stocks last year. Large-cap stocks were the main contributors to China’s stock rallies in the second half of last year, while the valuation premia in small cap stocks was compressed to near decade lows (Chart 4, bottom panel). Chart 4A Low Valuation Premia And More Policy Support May Further Lift Prices Of SMID-Caps Chart 5SMID-Caps Tend To Outperform Large-Caps In Late Business Cycle Historically, SMID-caps tend to outperform large-caps in the late cycle of an economic recovery (Chart 5). The spate of regulatory changes aimed at monopolistic behaviors in various sectors has curbed investors’ appetite for the industry leaders. In addition, the government’s increasing efforts to support small and medium corporates (SMEs) will help to shore up confidence in those companies. Therefore, small and medium caps will likely continue to outperform large-cap stocks this year.   Fiscal Support: How Much Room In 2H? The July Politburo meeting pledged more fiscal support for the economy later in 2021 and into 2022. We expect local government bond (LGB) issuance to accelerate: a 4.47 trillion RMB new local government bond issuance quota was approved for 2021, including 820 billion in general bonds and 3.65 trillion in special purpose bonds (SPBs). By end-August, 2.37 trillion new local government bonds had been issued, which was only 53% of the entire year’s goal. However, there are some constraints that will likely reduce the reflationary effects on the economy. First, the quota for LGB issuance approved by the National People’s Congress is 16% lower than last year, but the amount of LGBs maturing this year is 30% higher. Therefore, even though this year’s gross LGB issuance has kept pace with that of last year, more than half of the LGBs issued from January to August was used for debt repayment (Chart 6).  The move by local governments to use a large portion of their bond issuance quota to pay off existing debt resembles the situation in 2018 when a financial de-risking campaign encouraged local governments to reduce the stockpile of their leverage. As noted in last week’s report, infrastructure investment and the economy did not rebound in 2H2018, even though LGB issuance picked up (Chart 7). Chart 6More Than Half Of LGBs Issued This Year Has Been Used For Debt Repayment Chart 7Improvement In Infrastructure Investment Was Short-Lived In 2019 Even if we assume that local governments will use all of their remaining bond quota by year end, the gross monthly average in local government bond issuance will be around 580 billion, only slightly higher than in 2H 2020. Secondly, infrastructure investment is discouraged by stringent regulations to approve projects (including project assessment and debt repayment ability) and the accountability of local officials for project failures. Approvals for infrastructure projects remain at the lowest level since March last year (Chart 8). Finally, SPBs made up only about 15% of overall infrastructure spending in the past three years, while the majority came from public-private partnerships (PPP) financing, revenues from government-managed funds, government budgets and bank loans. Falling proceeds from land transfers have dragged down government-managed funds (Chart 9). In addition, government expenditures show no signs of a material increase (Chart 9, bottom panel). Chart 8Infrastructure Investment Will Remain Subdued Chart 9Government Expenditures Remain Muted As discussed in previous reports, local government bonds issuance only accounts for 12% of total social financing. As such, without a sizeable acceleration in bank loans, enhanced LGB issuance would not be enough to prompt a substantial increase in infrastructure investment growth. Our argument is underscored by the structural downshift in infrastructure investment since 2017 (Chart 7, top panel). Therefore, additional local government bond issuance this year will help to stabilize but not boost credit growth. August PMIs Confirm Slowing Economic Activity China's official PMIs eased further in August. The non-manufacturing index fell to contractionary territory of 47.5, below the expectation of a more muted 1.3-point decline to 52.0. Similarly, the manufacturing PMI eased by 0.3 points to 50.1, which is a hair above the 50 boom-bust line. Together, weakness in both sectors pushed down the composite index to 48.9 (Chart 10). Stringent restrictions designed to halt rising rates in COVID-19 infections explain much of the deterioration in China’s service-sector activity. The sector will likely rebound in September with the easing in infection levels (Chart 11). Chart 10PMIs Show Slowing Economic Activity Chart 11Lingering COVID Effects Curb Service-Sector Recovery In 2H21   Meanwhile, the construction PMI surprisedly rebounded sharply in August (Chart 10, bottom panel). However, investors should be cautious not to read too much into the idiosyncratic month-on-month moves suggested by the construction PMI. Instead, construction activity has moderated significantly and is set to slow further, hinting at plunged excavator sales and real estate investment in construction (Chart 12). Chart 12Construction Activity Is Unlikely To Pick Up Meaningfully This Year It is clear that China’s economy is losing momentum, but greater economic weakness will be needed for policymakers to stimulate meaningfully. Export Sector Remains A Bright Spot China’s exports remain robust. Export growth picked up in August from July on a year-over-year basis. Although the improvement in August reflects a base effect, exports in level reached a new high (Chart 13). Both skyrocketed exports container freight index and strong Korean exports suggest that global demand for Chinese manufacturing goods remains resilient (Chart 14). Even though manufacturing PMIs from developed markets have rolled over, they remain elevated and should continue to support China’s exports (Chart 15). Chart 13Chinese August Exports In Level Reached A New High Chart 14Exports Will Remain Robust In The Rest Of The Year... In contrast to resilient exports, China’s official PMI export new orders subindex has declined for five consecutive months. Even though falling PMI new export orders subindex heralds a slowing in exports growth, a reading of below the 50 boom-bust threshold in the former does not suggest a contraction in the growth rate of the latter. Furthermore, the month-over-month nature of PMI new export orders subindex tends to overstate the volatility in exports. The divergence between the PMI new export orders subindex and real export growth also occurred in 2018/19 during the height of the US-China trade war when export orders were volatile (Chart 16). Chart 15...And Will Continue To Benefit From Strong Global Demand Chart 16A Divergence Between PMI New Export Orders And Export Growth Regulatory Tightening In Real Estate Sector Stringent regulations in housing since the beginning of the year have started to cool the sector (Chart 17). However, home prices inflation in tier-one cities is still elevated (Chart 18). Thus, we expect the controls on housing and among property developers will remain in place for the next 6 to 12 months. Chart 17Housing Sector Is Cooling... Chart 18...But Housing Prices In First-Tier Cities Keep Rising At A Faster Rate Industrial Profits: Rising Prices, Falling Production China’s industrial profit growth remained solid in July despite the waning low base effect. Manufacturing producer prices continued to rise, offsetting weaker production growth (Chart 19). In addition, a low interest-rate environment helped to lift profits in the manufacturing sector by reducing debt servicing costs. While we expect weakening domestic demand and peaking producer prices to weigh on corporate profits in the rest of this year, profit growth is rolling over from a lofty height and will not likely drop sharply in the coming months (Chart 20). In addition, producer prices will likely remain at a historically high level in the next six months given robust global demand for raw materials and persistent global supply shortages. Chart 19Rising Prices And Low Interest Rates Helped To Offset Falling Industrial Production Chart 20Peaking Producer Prices Will Weigh On Corporate Profits Meanwhile, there is a large gap between the prices for producer goods and consumer goods, suggesting that manufacturers in mid-to-downstream industries have not been able to fully pass on rising input costs to domestic consumers (Chart 21). Profit growth continues to be disproportionally stronger in the upstream industrial producers than in the downstream industries, while the profit margin in the manufacturing sector remains much more muted (Chart 22).  Chart 21Inflation Passthrough From Manufacturers To Domestic Consumers Remains Limited Chart 22Profit Growth In Upstream Industries Still Outpaces Manufacturing Sector Table 1 Table 2 Footnotes   Market/Sector Recommendations Cyclical Investment Stance